Extract Office Content

目前已知问题

• 提取PPT:
  • 提取ppt中的内容时，会丢失带有公式的文本框
  • 提取的表格格式不全
  • PPT中的表格会提取为对应的excel文件，是否有更好的方式？
• 提取Word:
  • 表格位置不能与原文中一一对应

Use

1. Installextract_office_content

   $ pip install extract_office_content

2. Run by CLI.
   • Extract All office file's content.

     $ extract_office_content -h
     usage: extract_office_content [-h] [-img_dir SAVE_IMG_DIR] file_path
     
     positional arguments:
     file_path
     
     optional arguments:
     -h, --help            show this help message and exit
     -img_dir SAVE_IMG_DIR, --save_img_dir SAVE_IMG_DIR
     
     $ extract_office_content tests/test_files

   • Extract Word.

     $ extract_word -h
     usage: extract_word [-h] [-img_dir SAVE_IMG_DIR] word_path
     
     positional arguments:
     word_path
     
     optional arguments:
     -h, --help            show this help message and exit
     -img_dir SAVE_IMG_DIR, --save_img_dir SAVE_IMG_DIR
     
     $ extract_word tests/test_files/word_example.docx

   • Extract PPT.

     $ extract_ppt -h
     usage: extract_ppt [-h] [-img_dir SAVE_IMG_DIR] ppt_path
     
     positional arguments:
     ppt_path
     
     optional arguments:
     -h, --help            show this help message and exit
     -img_dir SAVE_IMG_DIR, --save_img_dir SAVE_IMG_DIR
     
     $ extract_ppt tests/test_files/ppt_example.pptx

   • Extract Excel.

     $ extract_excel -h
     usage: extract_excel [-h] [-f {markdown,html,latex,string}] [-o SAVE_IMG_DIR]
                         excel_path
     
     positional arguments:
     excel_path
     
     optional arguments:
     -h, --help            show this help message and exit
     -f {markdown,html,latex,string}, --output_format {markdown,html,latex,string}
     -o SAVE_IMG_DIR, --save_img_dir SAVE_IMG_DIR
     
     $ extract_excel tests/test_files/excel_example.xlsx

3. Run by python script.
   • Extract All.

     from pathlib import Path
     
     from extract_office_content import ExtractOfficeContent
     
     extracter = ExtractOfficeContent()
     file_list = list(Path('tests/test_files').iterdir())
     
     for file_path in file_list:
         res = extracter(file_path)
         print(res)

   • Extract Word.

     from extract_office_content import ExtractWord
     
     word_extract = ExtractWord()
     word_path = 'tests/test_files/word_example.docx'
     text = word_extract(word_path, "outputs/word")
     
     # or bytes
     with open(word_path, 'rb') as f:
         word_content = f.read()
     text = word_extract(word_content, "outputs/word")
     print(text)

   • Extract PPT.

     from pathlib import Path
     
     from extract_office_content import ExtractPPT
     
     ppt_extracter = ExtractPPT()
     
     ppt_path = 'tests/test_files/ppt_example.pptx'
     save_dir = 'outputs'
     save_img_dir = Path(save_dir) / Path(ppt_path).stem
     res = ppt_extracter(ppt_path, save_img_dir=str(save_img_dir))
     
     # or bytes
     with open(ppt_path, 'rb') as f:
         ppt_content = f.read()
     res = ppt_extracter(ppt_content, save_img_dir=str(save_img_dir))
     print(res)

   • Extract Excel.

     from extract_office_content import ExtractExcel
     
     excel_extract = ExtractExcel()
     
     excel_path = 'tests/test_files/excel_with_image.xlsx'
     res  = excel_extract(excel_path, out_format='markdown', save_img_dir='1')
     
     # or bytes
     with open(excel_path, 'rb') as f:
         excel_content = f.read()
     res  = excel_extract(excel_content, out_format='markdown', save_img_dir='1')
     print(res)

更新日志

• 2023-07-02 v0.0.6 update:
  • 统一提取word接口返回值为List，与其他统一
• 2023-06-17 v0.0.4 update:
  • 支持file-like object输入

Reference

• Pandas读取excel合并单元格的正确姿势（openpyxl合并单元格拆分并填充内容）
• python-docx2txt